WATER RELATIONS
OF WOODY PERENNIAL PLANT SPECIES
K. A. SHACKEL
Department of Plant Sciences/Pomology
University of California
Davis, CA, USA, 95616-8683
Abstract
Résumé
Aims: To describe the relation of various water status measures of woody
perennial plants (predawn and midday stem and leaf water potential), to
indices of physiological activity such as leaf conductance, vegetative growth
and fruit growth and composition.
Objectif : Décrire la relation des différentes mesures de l’état hydrique
des plantes ligneuses pérennes (potentiel hydrique foliaire et de tige de
base et à la mi-journée), aux indices d'activité physiologique comme la
conductance foliaire, la croissance végétative et la croissance du fruit et sa
composition.
Methods and results: Almonds were exposed to three levels of irrigation
over three years, and midday stem water potential (SWP) and leaf
conductance, collected at approximately weekly intervals, is reported for
the third year of the study. A strong linear increase in both leaf conductance
and trunk growth occurred with increasing SWP, and this relation was
consistent both within and between treatments. A similarly positive linear
relation was found between SWP and fruit size in pear, with a negative
relation between SWP and fruit soluble solids and fruit color. In grapevine,
SWP was found to be uniform across all lower canopy positions tested
(trunk, cordon and near the base of current year shoots) and positively
correlated to early season shoot growth even before irrigation treatments
were applied. Midday SWP was found to be more sensitive than midday
leaf water potential (LWP) for detecting treatment differences over the
course of the season, but both were well correlated to average seasonal
leaf conductance within and between irrigation treatments. Predawn SWP
and LWP were not as well correlated to average seasonal leaf conductance,
but the most important factor determining midday leaf conductance was
wind speed, indicating that grape leaf stomatal responses are quite sensitive
to this environmental factor.
Méthodes and résultats : Des amandes ont été exposées à trois niveaux
d'irrigation plus de trois ans, et le potentiel hydrique de tige à la mi-journée
et la conductance foliaire, collectés à intervalles quasi hebdomadaires, est
présenté pour la troisième année de l'étude. Il s’est produit une forte
augmentation linéaire tant dans la conductance foliaire que dans la croissance
du tronc avec l'augmentation du potentiel hydrique de tige, cette relation
était cohérente tant pendant qu'entre les traitements. Une relation linéaire
positive similaire a été trouvée entre le potentiel hydrique de tige et la taille
du fruit chez la poire, avec une relation négative entre le potentiel hydrique
de tige et les solides solubles du fruit et la couleur du fruit. Pour la vigne,
le potentiel hydrique de tige était constant pour toutes les positions de base
de la canopée (le tronc, le cordon et près de la base des tiges de l’année
courante) et positivement corrélé avec la croissance des pousses de début
de saison, même avant que les traitements d'irrigation n'aient été appliqués.
Le potentiel hydrique de tige à la mi-journée a été plus sensible que le
potentiel hydrique foliaire à la mi-journée pour détecter les différences de
traitement pendant la durée de la saison, mais les deux étaient bien corrélés
à la moyenne de la conductance foliaire saisonnière pendant et entre les
traitements d'irrigation. Le potentiel hydrique de tige de base et le potentiel
hydrique foliaire de base n'étaient pas aussi corrélés à la conductance foliaire
saisonnière moyenne, mais le facteur déterminant le plus important de la
conductance foliaire à la mi-journée était la vitesse du vent, indiquant que
les réponses des stomates de la feuille de vigne sont tout à fait sensibles
à ce facteur environnemental.
Conclusion: In a wide variety of woody crop species midday stem water
potential (SWP) has been found to be a valuable tool for quantifying the
degree of water stress experienced by the plant, and for understanding the
physiological responses of the plant to water limited conditions. In
grapevine, SWP detected irrigation differences over 1 month sooner than
midday leaf water potential when the number of vines used and the number
of samples taken were the same for both methods, and SWP had a higher
correlation to leaf conductance than predawn leaf or stem water potential.
Conclusion : Le potentiel hydrique de tige à la mi-journée est un outil
d’aide pour quantifier l’état hydrique de la plante et pour comprendre les
réponses physiologiques de la plante dans des conditions d’eau limitée.
Dans la vigne, le potentiel hydrique de tige a détecté des différences
d’irrigation un mois plus tôt que le potentiel hydrique foliaire à la mijournée lorsque le nombre de vignes utilisées et le nombre d'échantillons
pris était le même pour les deux méthodes et le potentiel hydrique foliaire
avait une corrélation plus forte à la conductance foliaire que le potentiel
hydrique foliaire ou de tige de base.
Significance and impact of study: SWP as a standard method for
quantifying water stress in grapevine and other crops will aid research in
the development of reliable management practices to improve crop
productivity and quality.
Key words: stem water potential, SWP, leaf water potential, LWP, predawn,
midday, leaf conductance, fruit growth, fruit quality
Signification et impact de l'étude Le potentiel hydrique de tige, comme
méthode standard pour évaluer quantitativement le stress hydrique de la
vigne et des autres cultures, aidera la recherche dans le développement de
pratiques fiables de gestion pour améliorer la productivité de la récolte et
la qualité.
Mots clés : potentiel hydrique de tige, SWP, potentiel hydrique foliaire,
LWP, conductance de feuille, croissance du fruit, qualité du fruit
A version of this work was presented at the VIth International Terroir Congress - Bordeaux - Montpellier 2006
manuscript received: 12th November 2006 - revised manuscript received: 7th August 2007
*Corresponding author: [email protected]
- 121 -
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
Shackel K. A.
water stress (e.g., the practice of Regulated Deficit
Irrigation, RDI) may be desirable for optimum
horticultural/economic productivity (e.g. Lampinen et al.,
1995, Shackel et al., 2000), and hence the assumption of
maintaining plants under non-stressed conditions would
not apply. Since the ultimate purpose of any management
practice designed to influence water availability (irrigation,
planting density, cover crop, etc.) is to obtain optimum
horticultural/economic productivity, I will propose that a
key property of any measure of water availability must
be that it is closely related to the plant responses that
determine horticultural/economic productivity. The plant
is a complex system, and fruit yield and quality are a final
result from the interaction of environmental conditions
with many physiological processes, such as vegetative
growth patterns, stomatal opening, photosynthetic rates,
root nutrient uptake, etc. Hence, as a first approximation,
any proposed measure of plant water availability, such as
plant water potential, should be closely related to these
processes.
INTRODUCTION
Understanding the physiological responses of plants
to water limited conditions is of general importance in
biology, as well as of particular importance to agricultural
science, and as such, this topic has been the subject of
much research and many excellent reviews (e.g., Bradford
and Hsiao, 1982; Jones and Tardieu, 1998). It is clear that
many, perhaps most, plants are designed to function over
a range of water availability, and that a number of short
and long term responses, often called « stress responses »
are exhibited by plants over this range. For the purpose
of this paper, I will use the approach suggested by Levett
(1980), with « stress » referring to an environmental
condition, which may be internal or external to the plant,
and « strain » as the impact of this condition on the
biological process of interest. I should admit at the outset
however, that I regard these and other related definitions
as provisional, since a clear and scientifically-based
terminology for describing the occurrence and
consequences of water stress in plants requires a thorough
understanding of the biological processes involved, and
for many of these processes our current understanding is
quite rudimentary. At the whole-organism level, water is
obtained from the soil by roots to re-supply water lost
from the leaves, and hence the term « water availability »
can typically be regarded as referring to soil water
availability, and « water limited conditions » as referring
to as any decrease in water availability that causes a strain
(i.e., change in biological activity) in the plant. In their
review, Bradford and Hsiao (1982) present a general
picture of the many plant stress responses that are known
to occur as water availability is decreased, with some
responses (i.e., reduction in growth rates) occurring first,
and others (stomatal closure, reduced photosynthesis)
occurring later, but one key question, and a focus of the
present paper, is the question of how we quantify the
decrease in water availability in the first place.
Since for any given experimental system (field site,
greenhouse, etc.) the points along the soil-plantatmosphere-continuum will be physically interdependent,
it is expected that any particular measure of plant water
availability within a site (soil-based, plant-based,
atmosphere-based) will probably be related to any other
measure of plant water availability at the same site. Hence,
if one of these measures is correlated to any given plant
physiological process or horticultural characteristic, then
we may expect that the other measures may also have
some correlation. The advantages of soil- and/or
atmosphere-based measures, and some plant-based
measures such as maximum daily shrinkage (Naor and
Cohen, 2003) are generally related to their practical
usefulness in irrigation scheduling (e.g. automation), but
it is reasonable to assume that plant-based measures of
water availability should be more directly related to plant
physiological responses than soil- or atmosphere-based
measures. A key physiological measure of water
availability within the plant is water potential, and although
some researchers have questioned the general relevance
of this measure to plant function (Sinclair and Ludlow,
1985) a few early studies in woody perennial crop species
(Garnier and Berger, 1985; McCutchan and Shackel,
1992) reported a close relation of water potential to both
irrigation regime and plant physiological activity, and
currently there are three alternative measures that have
been suggested: predawn water potential (PWP), midday
stem water potential (SWP) and midday leaf water
potential (LWP). A number of recent studies have
determined the relation of one or more these alternatives
to various physiological or horticultural characteristics,
and/or compared the sensitivity of these alternatives to
various irrigation practices. The sensitivity of a measure
in response to a change in irrigation practice is important
Many contrasting measures of plant water availability
have been suggested, each focusing at a different point
along the soil-plant-atmosphere-continuum. The
usefulness of any particular measure however, will depend
on the objective in making the measurement on the one
hand, and on the relation between the point being
measured and the desired plant response on the other. For
instance, many current irrigation practices are based on
direct measures of soil moisture, or indirect estimates
of soil available water through an estimation of
atmospheric evaporative demand (Ferreres and
Goldhamer, 1990). These approaches assume, among
other things, that the soil serves as a reservoir, and that
replacing the water lost from this reservoir is equivalent
to maintaining the plants under non-stressed, or at least
« minimally stressed » conditions. In the case of wine
grapes and other fruits however, moderate levels of plant
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
- 122 -
Water relations
if the method is to be used for irrigation management,
even if the method detects differences prior to any apparent
difference in physiological responses. Presumably the
most sensitive method for irrigation scheduling would
also be the most useful measure of plant water availability.
In review of the recent studies in deciduous orchard crops,
Naor (2006) concluded that LWP was not as closely related
to plant physiological processes as was PWP or SWP, and
that other indirect plant-based measures of water
availability, such as diurnal stem diameter changes, also
showed promise as practical tools for irrigation scheduling.
The following paper will present evidence in deciduous
trees and grapevine that the plant-based measure of midday
stem water potential (SWP) is closely related to many
important physiological processes, as well as key aspects
of fruit quality.
treatment and weekly or less often for the medium and
dry treatment. Irrigation amounts and timing were based
on maintaining contrasting levels of midday stem water
potential (SWP). A >50 year old commercial pear orchard
having trees of the « Bartlett » variety was irrigated with
under-tree impact sprinklers designed to give
approximately 100 % (WET), 60 % (MEDIUM) and
30 % (DRY1, DRY2) of tree water requirements for a
period of three years (1992 - 1994). In all years fruit size
and soluble solids was measured, but fruit color was only
measured in 1993. For the final year of this study (1994),
the start of irrigation on one of the two blocks of dry trees
was delayed (DRY2), based on the observation that trees
in this block required a longer period to deplete soil
moisture reserves. The field site for the grapevine study
was located on the campus of the University of California,
Davis. The plant material was own-rooted sixteen-yearold Pinot Noir vines trained using a bilateral « T » trellis
system with four cordons per trunk. Vines were planted
on 2.4 m spacing in East-West rows 3.7 m apart. Bud
break occurred on 15 March 2001 and flowering was on
26 April 2001. Two irrigation treatments were applied in
two large blocks using in-line drip irrigation beginning
22 May 2001 and continued through 24 August 2001:
non-irrigated and weekly irrigation to cumulative crop
evapotranspiration (ETc) levels. ETc was calculated using
the crop coefficient (Kc) values from Hanson et al. (1999),
and the reference evapotranspiration values (ET0)
calculated from a nearby weather station.
MATERIALS AND ANALYTICAL
AND EXPERIMENTAL METHODS
1. Physiological measurements
Values of water potential were measured with the
pressure chamber technique on leaves that were either
enclosed for more than 10 minutes to prevent transpiration
and allow equilibration with stem water potential (SWP,
Fulton et al., 2001), or leaves that were only enclosed
immediately before excision to prevent post-excision
desiccation (Turner and Long, 1980) to give leaf water
potential (LWP). In some cases, measurements were made
pre-dawn, also using both approaches. Unless otherwise
specified, LWP was measured on upper canopy, fully
exposed recently mature leaves, and SWP was measured
on lower canopy mature leaves that were attached to stems
that were close to the trunk or a main scaffold. Stomatal
conductance was measured on the abaxial (lower) surface
of upper canopy, recently mature sunlit leaves with a
steady-state porometer (Model 1600, LICOR, Lincoln,
NB), with minimal disturbance to the natural position
of the leaf. Trunk circumference was measured with a
steel tape, and converted to the equivalent cross-sectional
area of a circle as a relative measure of overall plant size.
When fruit was harvested, the number of fruits were
counted and the total fresh weight obtained to estimate
average fruit size, and other measures of fruit quality (color,
soluble solids) were measured on a sub sample using
standard laboratory techniques (colorimeter, refractometer).
RESULTS AND DISCUSSION
One practical advantage that has been reported for
midday SWP is the consistency of the measurement at
different positions within the plant. Presumably, this is
due to the fact that the xylem tissue is a very low resistance
pathway for water flow, and hence, for the rates of water
transport that are typically found during plant transpiration,
the gradient in SWP is not large within the stem portion
of the plant. For Pinot noir grapevine, we studied the
effect of position on SWP for positions near the trunk,
which should have the highest SWP, and for various
positions along the cordon and on primary and secondary
shoots (figure 1A). The correlation for SWP measured
at any two contrasting positions along the water flow
pathway across a wide range of SWP values was very
strong (figure 1B), with a mean difference of about
+0.01MPa (figure 1C). The random measurement error
calculated from these data (SD = 0.08MPa) is comparable
to that reported in other woody perennial species (SD =
0.05 - 0.07MPa, Fulton et al., 2001). The comparisons
shown in figure 1B are always for an upstream (X axis)
versus downstream (Y axis) position along the water flow
pathway (i.e., the direction of transpirational water flow
is always from position #1 to position #2), and hence it
is expected that the average difference would be slightly
2. Plant material and experimental design
Almond trees of the « Nonpareil » and « Carmel »
varieties were planted and grown in an experimental
orchard in Winters, CA, under three contrasting irrigation
regimes (WET, MEDUIM, DRY) for three years (1991
- 1993). Irrigation was supplied by under-tree low-volume
microsprinklers at twice-weekly intervals for the wet
- 123 -
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
Shackel K. A.
Figure 1 - (A) Diagram of the contrasting positions in cordon-trained grapevines that were used to evaluate the size
of within-plant differences that may occur in midday stem water potential (SWP) as a result of the water potential
gradients caused by transpiration. (B) Relation of SWP measured at a point upstream in the water flow pathway to
a point downstream in the pathway.
Line shown is the first principal component axis fit to the data (Y = 0.90*X - 0.054, regression r2 = 0.85***). (C) Histogram of the difference
between the X and Y values for each point shown in B. The overall mean ∀ SD of this difference was +0.0082∀0.081MPa.
negative, since the SWP should always decrease from
upstream to downstream. The fact that the difference was
slightly positive (+0.01MPa) presumably indicates that
there is no measurable SWP gradient within the accuracy
of these measurements. These data demonstrate a very
important practical advantage of SWP, namely that that
the measurement is essentially independent of the choice
of leaf on the major scaffold branches of the plant. This
result has also been found in a number of other woody
perennial species including prunes, almonds and pears
(data not shown).
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
Midday SWP (MPa)
Leaf conductance (mmol m-2 s-1)
Over the course of a three year study in almonds, clear
differences in SWP were exhibited for the three irrigation
treatments for most of the season in all years (figure 2A
shows the 1993 season), and there were corresponding
differences in leaf conductance from about mid-July
onward in 1993 (figure 2 B). For SWP, the tree-to-tree
variation was generally larger in the medium and dry
treatment compared to the wet treatment (note error bars
in figure 2A), but the ranking of trees within a treatment,
particularly the medium and dry treatments, was consistent
across dates and highly significant (ANOVA not shown).
Presumably, this indicated that there were individual trees
with more access to stored soil moisture reserves than
other trees in the same treatment. Since environmental
factors known to influence leaf conductance (air
temperature, light, etc.) changed from date to date, a
seasonal average leaf conductance value was calculated
for each individual tree, and this showed a strong and
positive correlation to the corresponding individual tree
seasonal average SWP (figure 3A). There was a similarly
strong correlation between the 3-year average SWP of
individual trees and tree size, as measured by the trunk
cross sectional area at the end of the 3-year study
(figure 3B). These data strongly support the hypothesis
that water deficits, as measured by SWP, are a key
determinant of plant physiological responses both at the
leaf level (conductance), and at the whole plant (growth)
level. Further, these data suggest that the dependence of
these responses is linear with SWP to a good
Figure 2 - Midday stem water potential (SWP, A) and
leaf conductance (B) during the 1993 growing season
in almonds under three contrasting levels of irrigation.
Also shown for reference (A) is the value of SWP expected for
fully irrigated almonds under the same midday conditions of air
temperature and relative humidity (baseline).
- 124 -
Trunk cross-sectional area (cm2)
Midday SWP (MPa)
Learf conductance (mmol m-2 s-1)
Water relations
Average fruit sige (g)
Figure 3 - (A) Relation of tree averaged leaf
conductance to tree averaged SWP in 1993 (data from
figure 2, linear r2 = 0.94***). (B) Relation of trunk
cross-sectional area after 3 years of growth (1991 1993) to tree averaged SWP over the same time period
for almonds under the three irrigation regimes shown
in figure 2 (linear r2 = 0.78***).
Fruit soluble solides (%)
Figure 4 - Midday stem water potential (SWP)
during the 1994 growing season in pears
under four contrasting levels of irrigation.
Figure 6 - Relation of pear fruit soluble solids at
harvest to tree averaged midday stem water potential
(SWP) during the main fruit growth period (June August) in 1994, as in figure 5 (linear r2 = 0.74***).
and more advanced coloration (yellow color, figure 7).
The effects of individual tree differences in SWP were
particularly clear for fruit size (figure 5), and showed that
there were some trees in the most severe dry treatment
(DRY 2) which had both SWP and fruit size values that
were comparable to some of the trees in the WET
treatment. Overall tree growth, as measured by pruning
weight, was also highly and positively correlated to SWP
(N = 47, r2 = 0.57***, data not shown).
Figure 5 - Relation of pear fruit size at harvest
to tree averaged midday stem water potential (SWP)
during the main fruit growth period (June - August)
in 1994 (linear r2 = 0.80***).
approximation. This contrasts with the widely held view
that one or both of these responses should exhibit a
threshold response to water deficits.
Hence it is clear that, as in almonds, reductions in
SWP had profound effects on many physiological
responses in pear trees, including a number of important
aspects of harvested fruit quality.
Similarly to young almond trees, well established pear
trees also showed a straightforward decline in SWP with
decreased irrigation and progressive separation between
treatments over time (figure 4). As also found in almond,
significant tree-to-tree variation was found, particularly
within the dry treatments, and there was a very strong
correlation between seasonal average SWP and many
measures of fruit quality. Lower SWP was associated with
smaller fruit size (figure 5), higher soluble solids (figure 6)
In May, prior to the application of any irrigation to a
block of Pinot noir grapevines, a survey indicated that
there were clear vine-to-vine differences in SWP, in some
cases with adjacent vines exhibiting differences on the
order of 0.5 MPa (figure 8). Differences in vine vigor
- 125 -
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
color (a)
Midday SWP (-MPa)
Shackel K. A.
Figure 8 - Spatial distribution of the midday stem
water potential (SWP) measured on individual vines
in May before any irrigation was applied. For clarity,
the SWP values are plotted as positive quantities.
Midday water potential (MPa)
Vine average shoot lenght (cm)
Figure 7 - Relation of pear fruit skin color (a value of
the l. a. b. scale) at harvest in 1993 (note difference in
year to figs. 5 and 6) to tree averaged midday stem
water potential (SWP) for the period of July - August
(linear r2 = 0.51***).
Figure 9 - Relation of shoot length at the end of May to
vine averaged midday stem water potential (SWP)
observed during May, prior to any irrigation, for a
subset of the vines shown in figure 8. Line is the linear
regression through the data (r2 = 0.38*).
Figure 10 - Midday stem water potential (SWP, top)
and leaf water potential (LWP, bottom) over the 2001
growing season for Pinot noir under irrigated
and non-irrigated conditions.
Each point is the mean of 6 vines, and error bars are ∀ 2SE, which
for this sample size is an approximate 95 % confidence interval.
Asterisks indicate a statistical difference between irrigated and nonirrigated vines.
were also apparent in this block, and the length of
randomly selected shoots on each vine by the end of May
had a significant positive correlation to vine SWP
(figure 9). Statistically significant differences in SWP
between irrigated and non-irrigated vines began to appear
in early June, and although the irrigation treatment effect
on LWP was of the same magnitude, LWP values were
more variable, and hence did not show statistically
significant differences until late July (figure 10). The daily
pattern in LWP, SWP and leaf conductance was followed
on 6 occasions in 2001, and both LWP and SWP showed
a similar daily pattern, with minimum values consistently
occurring in the afternoon period (2:00 - 3:00, figure 11).
As expected, there was also a progressive increase in
separation between irrigated and non-irrigated vines as
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
the season progressed in both SWP and LWP, but the
differences due to irrigation were more apparent in the
afternoon period than at other times of day, particularly
pre-dawn (figure 11). On any given day, the daily pattern
in stomatal conductance was similar for irrigated and
non-irrigated vines, and also showed a progressive
increase in separation between treatments as the season
progressed, but the daily pattern itself was not consistent
over the season. On some days there were relatively
constant values of conductance from mid morning to late
afternoon (figure 11, 9:00 - 17:00, June 12) whereas on
- 126 -
Midday SWP (-MPa)
Water relations
Figure 11 - Daily course of stem water potential (SWP), leaf water potential (LWP) and leaf conductance
in Pinot noir vines for a number of dates in the 2001 growing season under irrigated and non-irrigated conditions.
Each point is the mean of 6 vines, and error bars are ∀ 2SE, which for this sample size is an approximate 95 % confidence interval.
Table 1 - Results of stepwise regression analysis for midday leaf conductance as the dependent variable
and wind speed, air temperature solar radiation, air vapor pressure deficit (VPD), SWP and LWP
as the independent variables.
Values for conductance were averaged by vine (6-10 measurements per vine), but SWP and LWP values were only single measurements per
vine. These data are for all vines from all treatments taken on nine dates from June 12 through August 17, 2001.
temperature, SWP and radiation, all with a similar
influence (table 1).
other days a clear maximum value occurred, although the
time of maximal leaf conductance varied from late
morning on some days (11:00, July 20 and August 17) to
late afternoon on others (18:00, June 13). These data
suggest that grapevine leaf conductance is highly sensitive
to environmental factors as well as plant water status. A
stepwise linear regression was performed for midday leaf
conductance as the dependent variable and environmental
factors of radiation, temperature, vapor pressure deficit
(VPD) and wind speed, as well as SWP and LWP as
independent variables, and this analysis showed wind
speed as the most important factor, followed by
The relationship between SWP and conductance was
clearly influenced by wind, with windy days (wind speeds
> 3.5 m s-1) having generally low leaf conductance and
a lower regression slope of conductance on SWP than
low wind days (wind speeds < 2.8 m s-1, figure 12). On
low wind days, the average leaf conductance was
428.4 mmol m-2 s-1, compared to high wind days with
an average leaf conductance of 230.0 mmol m-2 s-1. In
order to compare the sensitivity of the alternative measures
of plant water deficits used in this study, an average water
- 127 -
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
Learf conductance (mmol m-2 s-1)
Learf conductance (mmol m-2 s-1)
Shackel K. A.
Figure 13 - Relation of vine average leaf conductance
to vine average leaf or stem water potential
during the daytime (LWP, SWP) or at predawn
(PDLWP, PDSWP), with averages taken over all days
and times shown in figure 11.
Figure 12 - Relation of midday leaf conductance to
midday stem water potential (SWP) for the Pinot noir
vines shown in figure 11, divided into environmental
conditions of high and low wind speed.
Each point is the mean of 3 - 5 values of conductance (fully exposed
leaves) and one value of SWP per vine.
Each point is the vine average value, and the lines are the regression
lines for each alternative measure of plant water status (r-square
value for each line is also indicated).
potential and leaf conductance was calculated for each
vine over all the dates and times shown in figure 11. Since
all vines were represented at each time point, a comparison
among these average values of leaf conductance will
represent a comparison among vines, averaged across a
wide range of environmental conditions. For these data
there was a significant and essentially linear relation
between leaf conductance and all of the alternative
measures of plant water status, but the correlation
coefficient of conductance to LWP and SWP was higher
than that to predawn LWP or SWP (figure 13). A close
and linear relation between midday LWP and SWP was
found by Williams and Trout (2005), and the values they
observed were over a somewhat wider range (SWP: -0.2
to -0.8, LWP: -0.5 to -1.1MPa) as found in our study, but
Williams and Trout (2005) also reported a much higher
range for predawn LWP (to -0.8MPa), with a non-linear
relation to midday LWP, probably indicating that much
higher levels in stress occurred in their most stressed
treatment as compared to our study. It is also interesting
to note that there was a systematic difference between
predawn LWP and SWP of about 0.08 MPa in our study,
presumably indicating that leaf, and hence plant,
transpiration may not be negligible at this time, as is
normally assumed. At midday, the difference between
LWP and SWP was typically 0.4 - 0.5 MPa (figure 11),
so attributing all of this predawn difference to transpiration
would imply that predawn transpiration was 16 - 20 %
of midday transpiration, which would be surprising. Hence
it is not clear whether all of the predawn difference was
due to transpiration or a combination of transpiration
effects and hydraulic conductance effects. In either case,
predawn leaf water potential has been used as a measure
of water stress/water availability in a number of viticultural
studies, and one advantage of the use of plant-based
measures for this is that the water stress experienced in
different studies can be compared on the same basis. For
this purpose, the data of Schultz (2003) was compared to
the data of the current study (figure 14). The range of
seasonal average stomatal conductance observed in the
present study corresponded to the upper range of the
conductances observed by Schultz (2003), but such
differences in stomatal conductance may be related to
differences in any number of environmental or cultural
conditions, for example lower air temperature leading to
lower stomatal conductance or higher soil fertility leading
to higher leaf nitrogen content and higher stomatal
conductance. For conductances over 150 mmol m-2 s-1
however, the equations used by Schultz (2003, figure 2)
systematically underestimated the conductances that he
observed (curves in figure 14), and for this range there
may have also been a relatively steep linear relation with
a low r2, similar to that shown in figure 13 and redrawn
in figure 14. In any case, it is clear that the level of stress
measured by Schultz (2003) far exceeded that found in
this study, and consistent with this interpretation, the low
levels of stomatal conductance found by Schultz (2003)
were not found in this study during daylight hours.
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)
CONCLUSION
In a wide variety of woody crop species we have
found midday stem water potential (SWP) to be a valuable
tool for quantifying the degree of water stress experienced
by the plant, and for understanding the physiological
responses of the plant to water limited conditions. The
primary criterion for any proposed measure of water sess
in plants should be that it is correlated with the well known
short- and long-term responses of plants to water limited
conditions, such as reductions in leaf conductance and
- 128 -
Water relations
Stomatal conductance (mmol m-2 s-1)
Ferreres E. and Goldhamer D.A., 1990. Deciduous Fruit and Nut
Trees. In: B.A. Stewart and D.R. Nielsen (eds) Irrigation
of Agricultural Crops. Agronomy Series (#30). Madison,
WI.
Garnier E. and Berger A., 1985. Testing water potential in peach
trees as an indicator of water stress. J. Hortic. Sci., 60,
47-56.
Hanson B., Schwankl L. and Fulton A., 1999. Scheduling
irrigations: when and how much water to apply. DANR
publication 3396, University of California, Davis, CA.
Jones H.G. and Tardieu F., 1998. Modelling water relations of
horticultural crops: a review. Scientia Horticulturae, 74,
21-46.
Lampinen B.D., Shackel K.A., Southwick S.M., Olson B.,
Yeager J.T. and Goldhamer D., 1995. Sensitivity of yield
and fruit quality of French prune to water deprivation at
different fruit growth stages. J. Amer. Soc. Hort. Sci., 120,
139-147.
Levitt J., 1980. Responses of plants to environmental stress. V1
(Chilling, freezing and temperature stress), V2 (Water,
radiation, salt and other stresses). Academic press, New
York.
McCutchan H. and Shackel K.A., 1992. Stem-water potential as
a sensitive indicator of water stress in prune trees (Prunus
domestica L. cv. French). J. Am. Soc. Hortic. Sci., 117, 607611.
Naor A., 2006. Irrigation scheduling and evaluation of tree water
status in deciduous orchards. Hort. Reviews, 111-165.
Naor A and Cohen S., 2003. Sensitivity and variability of maximum
trunk shrinkage, midday stem water potential, and
transpiration rate in response to withholding irrigation from
field-grown apple trees. HortScience, 38, 547-551.
Schultz H., 2003. Differences in hydraulic architecture account
for nearisohydric and anisohydric behaviour of two fieldgrown Vitis vinifera L. cultivars during drought. Plant Cell
Env., 26, 1393-1405
Shackel K.A., Lampinen B., Southwick S., Olson W., Sibbett S.,
Krueger W., Yeager J. and Goldhamer D., 2000. Deficit
irrigation in prunes: maintaining productivity with less
water. HortScience, 35, 1063-1066.
Sinclair T.R. and Ludlow M.M., 1985. Who taught plants
thermodynamics? The unfulfilled potential of plant water
potential. Aust. J. Plant Physiol., 12, 213-217.
Turner N.C. and Long M.J., 1980. Errors arising from rapid water
loss in the measurements of leaf water potential by the
pressure chamber technique. Aust. J. Plant Physiol., 7, 527537.
Williams L.E. and Trout T.J., 2005. Relationships among vineand soil-based measures of water status in a Thompson
seedless vineyard in response to high-frequency drip
irrigation. Am. J. Enol. Vitic., 56, 357-366.
Figure 14 - Relation of leaf conductance to predawn
leaf water potential for the data of figure 13
(UCD Pinot noir) and data reported
by Schultz (2003, figure 2) for Grenache and Syrah.
Also shown are the linear regression through the UCD data as in
figure 13, and the curvilinear relation for Grenache and Syrah given
in Schultz (2003, figure 2).
reductions in plant growth. For both of these responses
we have found very strong and essentially linear relations
with SWP in a number of crops, including grapevine, and
hence we suggest that SWP is a good candidate as a
standard measure of water stress in woody perennial plants.
We have also reported strong and essentially linear
relations of fruit quality attributes (size, soluble solids and
color) to SWP in pear, and hence propose that SWP should
be a useful tool to predict, and possibly manage, fruit
development in a wide range of woody perennial crop
plants.
Acknowledgements: This work was supported in part
by the California almond board, the California pear board, and
the UC Davis Young Scholars Program. The author
acknowledges the cooperation of Bill Biazi, Sunitha Gurusinghe,
Vivian Leung, Elizabeth Mitcham, Dave Ramos, and Jody
Smith.
BIBLIOGRAPHICAL REFERENCES
Bradford K.J. and Hsiao T.C., 1982. Physiological responses to
moderate water stress, p. 264-312. In: O.L. Lange, P.S.
Nobel, C.B. Osmond, and H. Ziegler (eds.) Physiological
ecology II. Water relations and carbon assimilation.
Encyclopedia of plant physiology. N.S. vol. 12B.
Fulton A., Buchner R., Olson B., Schwankl L., Gilles C.,
Bertagna N., Walton J. and Shackel K., 2001. Rapid
equilibration of leaf and stem water potential under field
conditions in almonds, walnuts and prunes. HortTechnology,
11, 609-615.
- 129 -
J. Int. Sci. Vigne Vin, 2007, 41, n°3, 121-129
©Vigne et Vin Publications Internationales (Bordeaux, France)